246 THE AMERICAN NATURALIST [Vol. LII 



pended diagram. Eelationsbips which have appeared 

 probable to the writer have been indicated by solid lines 

 connecting genera, possible relationships have been indi- 

 cated by dotted lines. The genera comprising the fami- 

 lies recognized by the Committee on Nomenclature as 

 belonging to a single family are enclosed by a heavy line. 

 Genera not recognized by the committee are enclosed in 

 dotted lines. 



The foregoing analysis would seem to indicate that the 

 grouping of genera by the Committee on Classification, 

 with some slight modifications possibly, represents fairly 

 well true phylogenetic relationships of the bacteria. The 

 exact boundaries of the families are of course of little im 

 portance providing the scheme of classification tends to 

 show relationships. 



